For eliminating noises in an image, a local averaging method for smoothing is usually applied, because this method can be easily executed by a software of a local spatial calculation. In the present large scale image processing system, an image is processed for the above smoothing in 32 msec.
The local averaging method dulls edges of a configuration and makes the image unclear as well as of low resolution. In order to solve this problem, various methods for smoothing without dulling edges have been proposed, such as the so-called Nagao method, local selecting method of pixels, hysteresis smoothing method, median filter method and E-filter method in frequency space. Of these methods the, median filter method is well known because of the high quality of its process result. However, the methods of smoothing without dulling edges, including the median filter method, need long process time and are difficult to be executed by hardware. These methods cannot be applied to processes that must be performed with high process speed.